An efficient memory data organization strategy for application-characteristic graph processing

Conclusion In this paper, we focus on the memory data organization strategy based on application operating characteristics and present GMDO, an efficient memory data organization strategy composed of GMDO-VR and GMDO-VI for traversal and iterative graph applications, respectively. Extensive experime...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Frontiers of Computer Science 2022-02, Vol.16 (1), p.161607, Article 161607
Hauptverfasser: FANG, Peng, WANG, Fang, SHI, Zhan, FENG, Dan, YI, Qianxu, XU, Xianghao, ZHANG, Yongxuan
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Conclusion In this paper, we focus on the memory data organization strategy based on application operating characteristics and present GMDO, an efficient memory data organization strategy composed of GMDO-VR and GMDO-VI for traversal and iterative graph applications, respectively. Extensive experiments show that GMDO exhibits remarkable performance compares with the representative schemes. For state-of-the-art graph systems, our proposed strategy can significantly improve the system acceleration ratio and increase the cache hit ratio.
ISSN:2095-2228
2095-2236
DOI:10.1007/s11704-020-0255-y